Heads and Splines and Software . . . Oh my!
With the advent of more powerful 3D software, the ability to create organic shapes
has improved much over the last few years. Artists are no longer restricted to simple
geometric objects, and are free to easily create flowing, more natural items. Of these,
the human head is one that is often chosen because of it's distinctiveness, and uses
in animation.
This article gives an overview of the process of creating a head out of splines. Specifically, I will discuss using 3D Studio MAX 1.2 and the Surface Tools plugin by Peter Watje. The surface tools plugin should be available for purchase by Digimation, and is a relatively cheap plugin. Note that there are other techniques for spline modeling, especially when using NURBS or other spline types. However, the process here should still yield some insight into the techniques I use, and be applicable to basic patch and spline modeling in a variety of other packages.
IMPORTANT: MAX 3.0 and later now has the surface tools suite built in. However, version 2.5 and later no longer allow you to weld the spline points! This is simply how splines now work in MAX. I am not sure why they changed it. So in cases below where it says to weld points, you will have to be happy simply 3d snapping them to the same location and leaving them separate. The surface modifier should still work. As a note, I'd now highly recommend using box modeling/meshsmooth modeling, instead of surface tools, even for modeling the face. In fact, the anime head, while it started as a surface tools model, was eventually converted to a low poly meshsmooth model the same as the body and finished up that way...which is why the final image here looks different than the ones in the images elsewhere on my page.
In fact I now have a Head Poly modeling tutorial on my site now as well that you can use instead!!!!
Finally, if you want to know more about how I did the hair, read the bottom of the body modeling tutorial on my site.
Note that this is not so much a tutorial as it is a walkthrough or overview! If you are looking for a tutorial, I'd highly recommend Bill Flemings book, "3D Creature Workshop" (ISBN: 1886801789). This book covers facial modeling in detail for a variety of packages, and may be more helpful.
Getting Started
The first thing you'll need to start with, is two reference photos of the head you wish to model.
These should be a front view, and a side/profile view. The photos should be as close
to a perfect front and side view as you can get. After scanning them in, make 2 boxes,
and apply the images to the box so that you can see it in a shaded viewport. The
snapshot on the right shows my reference images in a User view. (These were scanned
from an Evangelion art book, but I ended up changing my design at the end.)
The front image should be put on a box for front view. The side image on a box lines up
for a right or left viewport. The reason to use shaded boxes instead of just a backdrop
image, is that you can then zoom and pan in the front and side viewport as you work on the
splines. This gives more control over the modeling as you can zoom in and still have the
background line up to your splines as you work.
The next step is to start making the spline contours of your face in front view. Simply
trace the outline of half of the head and major features. Trace the contour around the eyes,
nose and mouth, as well as any detail you want around them. Then add cross sections
so that you have 3 or 4 sided patches. An easy way to do this in MAX is to turn on the "3D
Snap" button. Then in the Edit Spline 2 modifier, you can "create line" and create the
cross sections. Make sure you weld newly created points together.
Note that the anime head shown here isn't the best contour layout, since I was trying to keep it very low detail. I ended up fixing the splines later on as you can see in some of the later images. You probably should not use this image as a perfect example of how to start laying out your splines!
The image on the
right shows my head with a more detailed contour of splines. Actually, there are WAY too many
splines here, and I ended up deleting most of those as I worked on that head. You'll
probably want something with a level of detail in between these 2 image. (Look at the later images
in this article). But the picture of my head shows a little better layout.
Giving it Depth
After you are happy with the basic front view splines, the fun begins. The next step is to
pull your splines out into 3D. Using front and side views, pull put the each control point
until it matches the look of your profile reference photo. You'll probably want to start
with the splines at the middle of the head since these will be the ones matching that image.
The image on the right shows the anime head after being pulled into 3D.
With the anime head I had to drop following the profile too closely since the drawing in
side view didn't really match the way a 3D head would look. Even with real photos the side
view should be used as a starting reference only. After a certain point you may become aware
of perspective problems with the photos, and generally it is better simply to tweak it by
hand, using the photos as a guide only.
You can also start to pay attention to folds and curves. Pulling everything out to 3D is the
most critical, and time consuming part of the modeling. One can spend hours just tweaking
where a point is and playing with the bezier control handles. The amount of time you spend
making the head look correct here will affect how good your final image is. The image on
the right here is from my own realistic head model...still with too many splines.
After you are generally happy with the head, you should mirror it over to the other side.
This will often result in the need for more tweaking as the model can look quite different
when viewed in its full form.
At this point you can also model the back of the head and neck
area. This can be done before, or after mirroring and welding the two halves. Just make sure
you remove any extra splines at the joining edges, and weld all the vertices together. The image
at the right shows the back of the head and neck for my anime model. I tend to draw the back
of the head out in side view, and then pull it out sideways into 3D using front view.
As you work you will have an increasing number of confusing splines. This is especially
true as you get to the back and sides of the head. To help out, it is useful to create a
3D box. Intersect this box with the head in a shaded viewport. What will happen is the splines
behind it will be hidden. This is a nice way to see what you are doing since Edit Spline 2
doesn't have a way to hide control points or segments. Be warned however that you can still
select control points even though you can't see then. As I tweak the front of the face, I'll
usually have this "blocker" box up so the splines in the back don't confuse me. The image on
the right shows the final spline layout I used.
Finishing Up
I usually model a sack like shape that is connected to the inside of the lips for my
face models. This allows me to get a closeup of the mouth and what appears to be
the inside of the head. The tongue and teeth I usually create as separate polygon
objects. I texture this inside mouth sack with a dark red color. This image shows
the splines for the basic simple shape of the inside of the mouth.
Ears can be rather tricky to do, so I tend to model them separately. Then I merge
in the ear to the head model, welding and fixing splines as needed. There is a pretty
good discussion on ear modeling in Bill Flemings book listed at the top of this article.
The trick is in keeping only the amount of detail you need, while still maintaining the
proper shape.
The images to the right show the final spline model and rendering.
In this case I opted to model the eyebrows as separate poly models, but
in many cases it's just part of the texture map. I also created the
gums, teeth, tongue and eyelashes out of simple polygon objects as well.
For texture mapping I applied a cylindrical map around the head, and then did an unwrap texture to see the polygons. In photoshop simply paint on top of the features creating the textures. For photoreal heads, I'll utilize the scanned reference photos, cutting and pasting the major features like the ears and mouth, and distorting them to fit the unwrapped texture map. If you want to know a little more about how I did the hair, read the bottom of the body modeling tutorial on my site. I also now have a full blown hair guide under my 3d help pages as well.
A basic overview is below:
Remember also that I now have a Head Poly modeling tutorial on my site now as well that you can use instead of this one.
About the Author
Michael Comet is currently a Rigger/T.D. at Blue Sky Studios in New York.
Previously he was Video Team CG-Supervisor and Lead Rigger and a
3D Animator/Artist at
Big Idea in
Lombard, IL where he is worked on 3-2-1 Pengiuins and Veggie Tales.
Prior to that, he was lead animator at the video game company
Volition, Inc., where he animated most of the cinematic
sequences for Descent: Freespace, and headed up much of the realtime character animation and
cinematics R&D for their RPG title, "Summoner". He can be reached via email
at comet@comet-cartoons.com, and has a
personal homepage at: http://www.comet-cartoons.com/
which has more information and samples of his work
About the Sample Images
The head used was modeled with the Surface Tools plugin and 3D Studio MAX r1.2. The character is
part of a personal project I am working on.
This article, all images and character designs are Copyright 1998 Michael B. Comet All Rights Reserved.
This article may be reprinted for personal use only. It may not be packaged or sold in part or in whole, either alone or as part of another package. Unauthorized duplication is strictly prohibited. This article and related artwork, samples or text, are not to be copied onto other sites without prior written consent from the author. When in doubt, ask.